The present invention relates to automated data storage systems and more particularly, this invention relates to dynamically managing available storage within an automated data storage system.
Currently, data storage systems include a disk cache subsystem having different physical disk drive media (DDM). The differences in the DDM includes different physical drive types, capacities, manufacturers, rotational speeds, etc.
Due to the nature of such storage systems, if any number of DDMs are defective (e.g., suffering from faulty microcode, manufacturing problems, mechanical breakdown, etc.), the problem is not easily identifiable and/or avoidable once the DDM is installed in an automated data storage system in the field.
In order to mitigate or prevent lost data or productivity as a result of such failures, some users maintain storage devices in reserve, and initialize these devices into the system on an as-needed basis. However, precisely and accurately identifying a failing device and appropriate replacement continue to frustrate attempted improvements to data storage systems. In addition, in most systems, there is an anticipation that data storage demand will grow over time, and therefore it is appropriate to maintain reserved storage to manage that growth.
This is particularly troublesome where multiple users may be remotely sharing storage space and/or reserve storage devices across sysplexes. This is primarily because a given user in the shared storage system may not successfully recognize reserve storage devices which are in use by another user's system.
Thus, a major theme in improving data management strategy is simplification. One of the difficulties in realizing such improvements is storage management. While storage management via data facility storage management subsystem(s) (DFSMS) is still very advanced compared to most platforms, the number of steps required to add storage dynamically to a system is a concern. In particular, the multiplicity of required user interfaces and steps makes the process error-prone.
As mentioned above, it is typically the mainframe storage administrator's responsibility to manage a system that provides efficient use of resources. This involves maintaining storage groups such that they are tuned to application usage, as well as allowing for application/storage growth based on needs of the business. Some storage administrators plan to ensure storage is defined and made available to the different storage groups to accommodate future growth. Others maintain a set of “extra pre-defined volumes” and, on an as-needed basis, assign one or more to a particular storage group. When a new volume must be added, some customers estimate that it may take as long as hours or days between the time a request is made for more storage, to when it is fulfilled. In all cases, careful planning and numerous manual steps across various user interfaces are required, and each carries risk of error resulting in lost data.
In current data management schemes, the storage administrator must perform multiple steps using various interfaces in order to accomplish this task, according to the resident operating system (OS).
For example, when managing data on a system utilizing IBM® z/OS, these steps may include: identifying that a storage group needs additional capacity; identifying available volume(s) to add to the storage group; using an operator console, varying the volume(s) offline from all systems; submitting an “ICKDSF INIT” job to initialize the volume(s) (e.g. “volser,” “vtoc,” “sms,” etc.) for the target storage group; via a system display and search facility (SDSF), verifying that the job ran successfully; via interactive storage management facility (ISMF), updating a source control data set (SCDS) by adding “volser(s)” to the volume list of the target pool storage group; via ISMF, activating the SCDS; and varying the volume(s) online to all systems where needed.
Thus, increasing storage available to a given storage group currently includes performing a set of complex tasks to provide addressable system-managed disk storage for system and application use.